Adam Kan

I'm a second year undergrad at Carnegie Mellon University majoring in Computer Science and advised by Deepak Pathak and Kenneth Shaw. Previously, I was fortunate to be advised by Tom Funkhouser and Jeannette Bohg, and have worked with Jimmy Wu.

Email  /  CV  /  Google Scholar  /  LinkedIn  /  GitHub

Research

I'm interested in applying computer vision techniques for use in robots that help humans. My research experience is in building object detection models and visuomotor robot learning policies.

Robots can combine language-based planning and perception with the few-shot summarization capabilities of large language models (LLMs) to infer generalized user preferences that are broadly applicable to future interactions. Methodology is demonstrated in household cleanup tasks.

Utilizing current image segmentation abilities to detect the edges and centroids of cloths for bimanual unfolding tasks in sim. Our method represents grasp points by using angle value maps instead of spatial value maps to take advantage of generalizable physical characteristics of cloth unfolding tasks.

Modified template from here.